Every december Spotify makes you a “wrapped” playlist of the 100 songs you most frequently listened to that year giving a clear picture of your music taste that year. I’ve always found that my music taste is kind of variable with a lot of different genre’s so it would be interesting to see how my music taste has changed from 2016 till now. To compare my music taste from 2016 to 2020 i will analyze my Spotify wrapped 2016-2020 playlists using the following metrics: valence, danceability, energy, tempo, key and modality.
I would also be interesting to see if a classifier can keep the playlists apart form eachother.
I expect that my music will be average in valence since i do think i listen to happy and sad songs a fairly equal amount. Furthermore i expect to ahve listened to slighty more danceable and slightly less energetic music between 2016 and 2020 due to listening more to hiphop and electronic music and less to rock and pop. I expect the tempo to be all over the place as well as the key distribution. Modality is probably in a similar fashion to valence as i think major and minor mostly define if a song is happy or sad.
In addition i’d like to analyze my most frequently listened song and one favorite song or outlier from every playlist with the use of self similarity matrices, tempo, chroma and chordograms.
My Corpus is divided into 5 groups of spotify wrapped playlists representing 2016 to 2020:
The correlation between energy and danceability seems fairly similar in 2016 and 2018 indicating no correlation. Then 2017 and 2019 also look very similar with a positive correlation at around 50-60 danceability and 70-80 energy after which it usually goes down. In 2020 however, the correlation of energy and danceability seems negative.
My music from 2016-2020 has generally become more and more danceable every year shifting to the right until in 2020 almost all songs have a danceability higher than 0.4. My music seems to get slightly less energetic through the years. This can be explained by me listening more to hiphop and electronic music with high danceability but lower energy.
As mentioned in the previous tab you can see the music used for sleeping in the 2019 plot at the bottom left it has a clear low score on just about every variable.
2019 was excluded due to errors.
Overall the classifiers don’t work very well for my corpus but i expected as much. Since i do listen to a lot of different music the algorithms might have trouble finding distinct features for every year the different years. However, with limited features on the random forest classifier 2016, 2017 and 2020 seem to be working a lot better, almost classifying 50% correctly. The k-nearest neighbors classifier was trained on all of the features and the random forest trained only on the top half of the features from the feature importance plot.
Another thing worth noting is the correlation between 2016 and 2018 for both classifiers. The classifiers keep labeling 2016 as 2018 and vice versa. If we look at the the previous tab we can see where the mismatch originates from as two features energy and danceability seem to correlate in very similar ways in 2016 and 2018.
2016 Jesus of Suburbia - Green Day: a favorite because the song has lots of different sections and tempo changes. You can clearly see some of the different sections in the tempogram with tempo changes at 100, 200 and 400 seconds. However the bpm is way off starting at 300 BPM and almost reaching 400 at 400 seconds in. While the average BPM is around 147.
2017 Weightless part 1 - Marconi Union: an outlier because its sleep music and is not something i would listen to on the regular. As expected Spotify can’t really figure out the tempo as this song is just atmospheric sounds without a distinct way to measure tempo.
2018 Bohemian Rhapsody - Queen: a favorite because it has different sections and tempo changes. You can clearly see the segment change where the staccato piano comes in and the tempo changes at around 180 seconds in and when the guitars solo’s start at around 250 seconds in. Just as for every tempogram before the tempo seems way off with the song having an average bpm of 72, while the tempogram shows 2 lines one at around 100-110 bpm and one at around 300 bpm possibly indicating that spotify has trouble identifying the tempo of songs that have these kinds of tempo changes.
2019/2020 Build God, Then We’ll Talk - Panic! At The Disco: a favorite because it has different sections, tempo and time signature changes making it a very unique song. At 50 seconds in, the time signature changes to a 3/4 and then reverts back to 4/4 after a small section then another time signature change just after the 1 minute mark explains the small gap in the line. Spotify correctly finds the tempo change at 130 seconds. Once again the BPM is totally wrong with this song having an average tempo of 124.
Outliers:
2016 Jesus of Suburbia - Green Day: a favorite because the song has lots of different sections and tempo changes. You can see the different sections a little bit shifting from A major to Ab Major to A minor and B minor then going back to and eventually going back to B minor in the end.
2017 Weightless part 1 - Marconi Union: an outlier because its sleep music and is not something you would listen to on the regular. As expected you cant really make up a clear structure since the sounds are so atmospheric.
2018 Bohemian Rhapsody - Queen: a favorite because it has different sections and tempo changes. Spotify finds it hard to make up the chord structure but does map out the solo section starting at 210 seconds in pretty well.
2019/2020 Build God, Then We’ll Talk - Panic! At The Disco: a favorite because it has different sections, tempo and time signature changes making it a very unique song. As complicated as this song is structured Spotify seems to pick up on a lot of different chords but it’s hard to see patterns emerge from it.
I expected the valence scores in my playlist to be consistently average. It appears however, that from 2016 to 2018 i started listening to happier music, after that the valence score reverted back to around 0.5 indicating that in 2019 and 2020 i mostly listened to music that’s in between happy or sad.
My music from 2016-2020 has generally become more and more danceable every year shifting to the right until in 2020 almost all songs have a danceability higher than 0.4. My music seems to get slightly less energetic through the years. This can be explained by me listening more to hiphop and electronic music with high danceability but lower energy. It’s interesting that the correlations between the 2016 and 2018 playlists and the correlation between 2017 and 2019 are fairly similar. This indicates that regarding danceability and energy my music taste for those years had been pretty similar. 2020 was more of an odd one out as the correlation looks nothing like the others, indicationg a distinct difference in danceability and energy.
The classifiers had trouble distinguishing the Spotify Wrapped playlists from eachother. Probably meaning my music taste did not change that much. In 2020 however the music seemed to be distinct enough to correctly classify 50% of the playlist as coming from wrapped 2020. Songs from the 2016 and 2018 playlists kept being classified as eachother which indicates my music taste in these years must’ve been fairly similar. This is also supported by the correlation in the Danceability and Energy plot which is fairly similar in 2016 and 2018. The classifiers don’t seem to show a correlation between 2017 and 2019 as much as was apparent from the Danceability and Energy plot.
I can conclude that i have a fairly clear preference for songs with a tempo around 100 BPM which is interesting since it doesn’t conform with Moelants(2002) which suggests that humans prefer tempi around 120-130 BPM.
The key frequency is fairly distributed across all years and doesn’t change that much other than flattening out a bit in later years. As for the mode it is indeed fairly similar to how valence progressed as expected. The mode i evenly distributed between major and minor over all years except for a peak in major mode in 2018.